Utilities for training deep learning models in pytorch and scikit-learn, developed by Xtract AI.
The xt-training package has been built by extending the core functionality of pytorch. Core pytorch components (such as datasets/dataloaders, modules, loss functions, optimizers, and schedulers) are defined by the user in the same way, with xt-training providing a layer on top to provide more repeatable, efficient, and better organized code, logging, and experiment management.
The easiest way to install xt-training is using pip. Assuming you have python (>= 3.5) and pip install on your system, simply run:
pip install xt-training
To upgrade an existing installation, run:
pip install --upgrade xt-training
To install from source, first clone the xt-training repo:
git clone https://github.com/XtractTech/xt-training.git
Then use pip to install the module:
pip install ./xt-training
Next, see Getting Started to learn how to make the most of xt-training.
See the Github repo pull requests for ongoing work.
If you wish to make a contribution to the code (or the docs), PRs are more than welcome.
See the Github repo issues page for help.
TODO: add a license.